SShogo Bench

GPT-6 Sol vs Grok 4.5

GPT-6 Sol leads Grok 4.5 by 14.2 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. GPT-6 Sol wins 18 and Grok 4.5 wins 1 of 19 shared benchmarks. Grok 4.5 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGPT-6 SolGrok 4.5
Agentic coding7254
Long-horizon autonomy7659
Knowledge-work agents4842
Computer use5436
Security & context learning5948

Shared benchmarks

BenchmarkGPT-6 SolGrok 4.5
APEX-Agentsindex54.3%56.2%
Vending-Bench 2index$14,428$3,887
ARC-AGI95.5%87.2%
ARC-AGI-289.6%52.6%
Chess Puzzles56.0%35.0%
CritPt30.9%15.4%
DTBench97.3%96.5%
FrontierCode49.3%42.4%
FrontierMath Tier 490.0%24.4%
FrontierMath Tiers 1-389.8%57.2%
GPQA Diamond94.3%93.4%
LMCA59.1%45.2%
MirrorCode48.9%32.2%
OTIS Mock AIME100.0%97.8%
ProofBench83.0%31.0%
SciCode57.6%55.0%
SimpleBench73.1%70.0%
SimpleQA Verified60.7%48.3%
WebDev Arena16891555

Specs and price

GPT-6 SolGrok 4.5
Agent Index63.6 (#13)49.4 (#36)
Agent $/M$0.92$0.81
Input / output $/M$2.00 / $10.00$2.00 / $6.00
Context1.1M500K
ReleasedSep 22, 2026Jul 8, 2026
WeightsClosedClosed
API hosts54

Don’t pick one. Route.

Shogo can run GPT-6 Sol and Grok 4.5 side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo